JP5454266B2 - Node, communication program, and communication method - Google Patents

Description

  The present invention relates to a node, a communication program, and a communication method for controlling communication between nodes in a network in which first and second communications are mixed.

  In recent years, modularization for each function has been performed in machine control devices such as automobiles, industrial robots, and humanoid robots. In addition, with the advent of sensors having a network function, it has become common for terminal devices (modules) mounted on machine control devices to form a network.

  In a machine control apparatus including such a network, generally, real-time communication and non-real-time communication are mixed. Real-time communication is communication that guarantees that data transmission / reception is completed at certain time intervals. As real-time communication, there is communication in which data with a relatively small amount of data (several tens to hundreds of bytes) is transmitted and received periodically (several milliseconds) with extremely low delay for feedback control of the machine control device.

  As real-time communication, in USB 2.0, IEEE 1394, etc., in order to transfer audio and moving images without delay, a transfer opportunity is always obtained once in an extremely short period (several tens to hundreds of microseconds). There is isochronous transfer that transfers several kilobytes of data. USB is an abbreviation for “Universal Serial Bus”. IEEE is an abbreviation for “Institut of Electrical and Electronic Engineers”.

  In contrast to real-time communication, non-real-time communication is communication that does not require real-time performance. Non-real-time communication includes, for example, communication that can withstand relatively delay such as distribution of application modification programs and file transfer but requires high throughput.

  On the other hand, in the machine control device including the network as described above, a general-purpose network (for example, Ethernet (registration) that includes an aggregation / distribution device such as a switch in the route from the viewpoint of flexibility of system configuration, expandability, and ease of construction. Trademark)) is desired.

  However, in the switch, when a plurality of inputs transferred to the same output terminal exceed the throughput of the output terminal, the communication queue in the switch grows. For this reason, in a network including a switch in the middle of a route, growth of a communication queue may cause an increase in communication delay and a packet loss, and it is difficult to appropriately execute real-time communication.

  Conventionally, there has been a technique for temporally distinguishing between execution periods of real-time communication and non-real-time communication using the periodicity of real-time communication. Specifically, the relay device that relays between the terminal devices, when relaying the isodata packet, if a plurality of isodata packets compete, by outputting a plurality of isodata packets collectively, The period during which the output of non-iso data packets is suppressed is shortened.

  There is also a technique for reliably transmitting media data such as voice and moving images. Specifically, in the data transfer method, a resource (for example, network bandwidth) necessary for executing data transfer between information transmission apparatuses in a network is reserved in advance, and a packet is transmitted using the reserved resource. Perform routing processing.

JP 2006-279852 A JP 2001-268105 A

  However, in the above-described prior art, in a network in which terminal devices are connected via a plurality of switches, in an environment where real-time communication and non-real-time communication are mixed, non-real-time communication on a path connecting the switches is performed. It is difficult to control the flow rate. For this reason, there is a problem in that it is difficult to eliminate a communication collision on a path connecting between switches, and it is difficult to guarantee real-time performance of real-time communication.

  On the other hand, if a system configuration that connects terminal devices via a plurality of switches is not adopted, the maximum number of terminal devices that can be connected to the system is limited, and the flexibility and expandability of the system configuration is reduced. is there.

  In the above-described prior art, the execution period of real-time communication and non-real-time communication is distinguished in terms of time, and during the period when real-time communication is performed, non-real-time communication is stopped and collision with real-time communication occurs. It is also possible to avoid this. However, if all non-real-time communication is stopped during real-time communication, there is a problem that the throughput of the entire network is reduced.

  In order to solve the above-described problems caused by the prior art, the present invention provides a node, a communication program, and a communication program capable of avoiding a collision between the first and second communications in a local network in which the first and second communications are mixed. An object is to provide a communication method.

  In order to solve the above-described problem and achieve the object, a disclosed node includes a first node group including a first node and a second node, and a second node including a third node and a fourth node. A first node that executes first communication and second communication with a node group via a network, the first node being performed between the second node and the fourth node; Data relating to communication 2 is received from the second node, a communication path through the network is set between the first node and the third node, and based on the transmission capability of the network The time required for transmission of the transmission target data sequentially determined from the start point of the transmission process periodically executed to transmit the data sequentially determined as the transmission target from the received data. Calculate Determining whether or not the time when the required time has elapsed from the current time coincides with the start time of the first communication periodically executed between the second node and the fourth node; It is a requirement to determine the data to be transmitted from the received data based on the determined determination result.

  According to this node, the communication program, and the communication method, there is an effect that it is possible to avoid a collision between the first and second communications in a local network in which the first and second communications are mixed.

It is explanatory drawing which shows one Example of this communication method. It is explanatory drawing which shows the outline | summary of the example of determination of the data used as transmission object. 1 is a system configuration diagram (part 1) of a network system. FIG. It is a block diagram which shows an example of the hardware constitutions of a node. It is explanatory drawing which shows an example of the memory content of an address table. It is explanatory drawing which shows an example of the memory content of a group identification table. It is a block diagram which shows the functional structure of a relay node. It is explanatory drawing which shows an example of the data structure of non-real time data. It is explanatory drawing (the 1) which shows the specific example of a destination / transfer destination correspondence table. It is explanatory drawing which shows an example of the memory content of a queue corresponding | compatible table. It is explanatory drawing which shows an example of the memory content of an allocation flow volume / transmitted data amount corresponding | compatible table. It is explanatory drawing which shows an example of a coincidence determination with the period start time of real-time communication. It is explanatory drawing (the 1) which shows an example of the memory content of a real-time communication allocation flow rate table. It is explanatory drawing (the 1) which shows an example of the memory content of a transmission possible destination list | wrist. It is a block diagram which shows the functional structure of a group node. It is explanatory drawing (the 2) which shows the specific example of a destination / transfer destination correspondence table. It is explanatory drawing (the 2) which shows an example of the memory content of a real-time communication allocation flow rate table. It is explanatory drawing (the 2) which shows an example of the memory content of a transmission possible destination list | wrist. It is a flowchart which shows an example of the communication processing procedure of a relay node. It is a flowchart which shows an example of the specific process sequence of the period start time correction process of step S1913. It is a flowchart (the 1) which shows an example of the specific process sequence of the transfer control process of step S1914. It is a flowchart (the 2) which shows an example of the specific process sequence of the transfer control process of step S1914. It is a flowchart which shows an example of the communication processing procedure of a group node. It is a flowchart (the 1) which shows an example of the specific process sequence of the transmission control process of step S2308. It is a flowchart (the 2) which shows an example of the specific process sequence of the transmission control process of step S2208. It is a flowchart which shows an example of the flow rate allocation request | requirement processing procedure of real-time communication. It is a flowchart (the 1) which shows an example of the flow rate allocation response process procedure of real-time communication. It is a flowchart (the 2) which shows an example of the flow rate allocation response process procedure of real-time communication. It is a system configuration | structure figure (the 2) of a network system.

  Exemplary embodiments of a node, a communication program, and a communication method according to the present invention will be explained below in detail with reference to the accompanying drawings.

(One example of this communication method)
First, an example of the communication method according to the embodiment will be described. In this communication method, communication between nodes is controlled so that the first communication and the second communication do not collide in a local network in which the first and second communications are mixed.

  Here, the first communication is communication periodically executed between any one of the nodes included in the local network. For example, the first communication is real-time communication for feedback control of the machine control device, isochronous transfer for transferring voice, moving images, or the like without delay. In this specification, isochronous transfer is also treated as real-time communication from the viewpoint of ensuring that data transmission / reception is completed at certain time intervals.

  The second communication is a communication that is irregularly executed between any nodes in the node group included in the local network. For example, the second communication is a non-real-time communication that can withstand a relatively long delay such as distribution of a modification program of an application or file transfer but requires high throughput.

  FIG. 1 is an explanatory diagram showing an embodiment of this communication method. In FIG. 1, a local network 100 is configured to include node groups N1 to N6 and switches SW1 and SW2.

  (1) Node groups directly connected to the individual switches SW1 and SW2 are grouped. Here, a group A is formed by the node groups N1 to N3 directly connected to the switch SW1, and a group B is formed by the node groups N4 to N6 directly connected to the switch SW2.

  (2) A relay node for relaying second communication between nodes belonging to different groups A and B is installed for each of groups A and B. Here, node N1 is installed as a relay node of group A, and node N4 is installed as a relay node of group B. Hereinafter, an embodiment of the present communication method will be described by focusing on the relay node N1 of the group A among the relay nodes N1 and N4.

  (3) The relay node N1 sets a virtual link (hereinafter referred to as “virtual link VL”) that connects between the relay nodes N1 and N4 using a virtual network technology. The virtual link VL is used to relay second communication between nodes belonging to different groups A and B (for example, between the nodes N2 and N5).

  Specifically, for example, the relay node N1 receives data related to the second communication with respect to the node N5 of the group B from the node N2 of the group A, and transmits the data to the relay node N4 via the virtual link VL. The data transmitted from the relay node N1 to the relay node N4 is transmitted from the relay node N4 to the node N5.

  Here, the relay node N1 sequentially determines the received data group related to the second communication as a transmission target in the following (6), and stores it in the buffer of the relay node N1. As a result, when the relay node N1 starts the transmission task, it sends the data in the buffer to the local network 100 in the order of storage.

  (4) Based on the transmission capability of the local network 100, the relay node N1 calculates the time t required to transmit the data sequence sequentially determined as the transmission target from the start time of the periodically executed transmission task to the present time. To do. That is, the relay node N1 estimates the time taken to send the data string currently stored in the buffer to the local network 100.

  (5) The relay node N1 is periodically executed when the required time t has elapsed from the present time between any of the plurality of node groups N1 to N6 (for example, between the nodes N2 and N5). It is determined whether or not it coincides with the start time of communication. Here, all the nodes N1 to N6 are synchronized at regular time intervals.

  (6) The relay node N1 determines data to be transmitted from the received data group related to the second communication based on the determined determination result. An example of determining data to be transmitted will be described later with reference to FIG.

  (7) The relay node N1 transmits data sequentially determined as transmission targets to the relay node N4 via the virtual link VL. Specifically, the relay node N1 activates the transmission task, reads the data strings in the buffer in the order of storage, and sends them to the local network 100.

  FIG. 2 is an explanatory diagram showing an outline of an example of determining data to be transmitted. Here, an example of determining data to be transmitted in the relay node N1 will be described. In FIG. 2, data D1 to D12 are a data group related to the second communication received at the relay node N1. Further, the node N5 in the group B will be described as an example of the communication source of the first communication executed between any of the nodes N1 to N6.

  In FIG. 2, a data string 210 represents data strings Da to Dj stored in the buffer of the node N5. These data strings Da to Dj are read from the buffer in the order of storage (data Da → data Db →...) And sent to the local network 100 when the transmission task of the node N5 is activated. As a result, data Da, Dc, De, Dg, Di related to the first communication is transmitted to the local network 100 at regular intervals.

  In this communication method, in the above (6), when it coincides with the start time of the first communication, the data (for example, data D1) of the route that does not collide with the route of the first communication is selected from the data groups D1 to D12. The transmission target is determined (data string 220). On the other hand, in the above (6), when the start time of the first communication is not coincident, arbitrary data (for example, data D2) is determined as a transmission target from the data group (data string 220).

  As described above, in this communication method, the data of the path that does not collide with the path of the first communication is sequentially determined as the transmission target, thereby avoiding the collision of the first and second communications. At this time, it is not necessary to control the transmission timing in accordance with the start time of the first communication by shaping the data sequence 120 in advance so as not to collide with the first communication and storing it in the buffer.

(System configuration of network system 300)
Next, a system configuration of the network system 300 according to the embodiment will be described. FIG. 3 is a system configuration diagram (part 1) of the network system. In the network system 300, nodes N1 to N12 and switches SW1 to SW5 are communicably connected via a network 310 such as a LAN (Local Area Network).

  Here, the nodes N1 to N12 are communication devices having a communication function (real-time communication, non-real-time communication). For example, the nodes N1 to N12 are ECUs (Electric Control Units), sensors, and actuators mounted on automobiles and robots. is there.

  The switches SW1 to SW5 are relay devices that have a function of relaying data between nodes. When the switches SW1 to SW5 receive the data, for example, the switches SW1 to SW5 transfer the data to any one of the nodes N1 to N12 according to the physical address specified as the destination.

  In the network system 300, groups G1 to G4 are formed by node groups that are directly connected to the individual switches SW1 to SW4. Specifically, a group G1 is formed by the node groups N1 to N3 directly connected to the switch SW1, and a group G2 is formed by the node groups N4 to N6 directly connected to the switch SW2. A group G3 is formed by the node groups N7 to N9 directly connected to the switch SW3, and a group G4 is formed by the node groups N10 to N12 directly connected to the switch SW4.

  Furthermore, in the network system 300, relay nodes N1, N4, N7, and N10 that relay non-real-time communication between groups are provided for each of the groups G1 to G4. In the network system 300, virtual links VL that connect the relay nodes are set.

  In the following description, “virtual link VLx-y” represents a virtual link from the relay node of group Gx to the relay node of group Gy. However, “virtual link VLx-y” and “virtual link VLy-x” represent the same virtual link.

  Specifically, virtual links VL1-2, VL1-3, and VL1-4 are set between the relay node N1 of the group G1 and the relay nodes N4, N7, and N10 of the groups G2 to G4. Virtual links VL2-3 and VL2-4 are set between the relay node N4 of the group G2 and the relay nodes N7 and N10 of the groups G3 and G4. A virtual link VL3-4 is set between the relay node N7 of the group G3 and the relay node N10 of the group G4.

(Hardware configuration of nodes N1 to N12)
Next, the hardware configuration of the nodes N1 to N12 illustrated in FIG. 3 will be described. FIG. 4 is a block diagram illustrating an example of a hardware configuration of a node. 4, nodes N1 to N12 include a CPU (Central Processing Unit) 401, a ROM (Read-Only Memory) 402, a RAM (Random Access Memory) 403, and an I / F (Interface) 404. Yes. Each component is connected by a bus 400.

  Here, the CPU 401 governs overall control of each of the nodes N1 to N12. The ROM 402 stores programs such as a boot program. The RAM 403 is used as a work area for the CPU 401. The I / F 404 is connected to a network 310 such as a LAN through a communication line, and is connected to other devices via the network 310. The I / F 404 controls an internal interface with the network 310 and controls input / output of data from an external device.

  The nodes N1 to N12 may include a magnetic disk drive, a magnetic disk, an optical disk drive, an optical disk, and the like in addition to the above components. Here, the magnetic disk drive controls the reading / writing of the data with respect to a magnetic disk according to control of CPU401.

  The magnetic disk stores data written under the control of the magnetic disk drive. The optical disc drive controls reading / writing of data with respect to the optical disc according to the control of the CPU 401. The optical disk stores data written under the control of the optical disk drive, and allows the computer to read data stored on the optical disk.

(Contents stored in various tables)
Next, the contents stored in the various tables 500 and 600 used by the nodes N1 to N12 will be described. The various tables 500 and 600 are stored in a storage device such as the ROM 402 and the RAM 403 shown in FIG. In the following description, “nodes N1 to N12” are simply expressed as “node N” unless otherwise specified.

<Storage contents of address table>
FIG. 5 is an explanatory diagram showing an example of the contents stored in the address table. In FIG. 5, the address table 500 has a node name, a system address, a group address, a physical address, and relay node information for each of the nodes N1 to N12.

  The node name is a name of the nodes N1 to N12 used for explanation in this specification. The system address is an address of each node N1 to N12 that is uniquely determined in the entire network system 300. The physical address is an address unique to the I / F 404 connected to the network 310. The physical address is, for example, an Ethernet (registered trademark) MAC (Media Access Control address) address.

  The group address is an address of each node N1 to N12 uniquely determined by each group G1 to G4 in the network system 300. Here, the first to third sets of numbers (for example, 172.xxx.1.1) among the four sets of numbers separated by the dot “.” Of the group address are group identification IDs for identifying the groups G1 to G4. is there. The fourth set of numbers (for example, 1) is a node identification ID for identifying the node N in the group.

  The relay node information is information for identifying the node N that functions as a relay node. Here, “Yes” is set in the relay node information of the node N functioning as the relay node, and “No” is set in the relay node information of the node N functioning as the group node. Note that one (or more) relay node is always determined from each of the groups G1 to G4 based on preset information or a selection algorithm.

<Storage contents of group identification table>
FIG. 6 is an explanatory diagram of an example of the contents stored in the group identification table. In FIG. 6, the group identification table 600 stores a group name and a group identification ID in association with each other for each of the groups G1 to G4.

  The group name is a name of the groups G1 to G4 used for explanation in this specification. The group identification ID is an identifier for identifying each group G1 to G4. The group identification ID is, for example, a network part of an IP (Internet Protocol) address. Here, the upper part of the group address (for example, 172.xxx.1) corresponds to the group identification ID.

  In the following description, among the nodes N1 to N12 in the network 310, the node N that relays non-real-time communication is denoted as “relay node JN”, and the remaining nodes N excluding the relay node JN are denoted as “group nodes”. It is written as “GN”. Here, the functional configuration of the relay node JN will be described.

(Functional configuration of relay node JN)
FIG. 7 is a block diagram illustrating a functional configuration of the relay node. In FIG. 7, the relay node JN includes a receiving unit 701, a setting unit 702, an associating unit 703, a transmitting unit 704, a detecting unit 705, an assigning unit 706, a first calculating unit 707, and a determining unit 708. And a determination unit 709, a second calculation unit 710, an extraction unit 711, and a determination unit 712.

  Specifically, each function unit (reception unit 701 to determination unit 712), for example, causes the CPU 401 to execute a program stored in a storage device such as the ROM 402 and the RAM 403 illustrated in FIG. The function is realized by F404. Note that the processing results of each functional unit (reception unit 701 to determination unit 712) are stored in a storage device such as the RAM 403 unless otherwise specified.

  In the following description, when a specific example of each functional unit (reception unit 701 to determination unit 712) of the relay node JN is described, the node N1 in the network system 300 is assumed as the relay node JN. Further, real-time communication and non-real-time communication between nodes N will be described as examples of first and second communication between nodes N that are mixed in the network system 300.

  First, the receiving unit 701 receives various data. The various data includes, for example, data related to non-real time communication (hereinafter referred to as “non-real time data”), a synchronization packet, a flow rate allocation request and a flow rate allocation response for real time communication, and the like. A detailed description of the synchronization packet, the flow rate allocation request, and the flow rate allocation response will be described later.

  Specifically, for example, the receiving unit 701 receives non-real time data for a group node GN included in another node N group from a group node GN included in one node N group among the plurality of node N groups. . Here, the node N group is, for example, a set of nodes N directly connected to the individual switches SW1 to SW4 illustrated in FIG. One node N group is a node N group including its own node N. Here, the data structure of non-real time data will be described.

  FIG. 8 is an explanatory diagram showing an example of the data structure of non-real time data. In FIG. 8, the non-real time 800 has a header area and a data area. In the header area, a physical destination address, a logical destination address, and a data amount are set. Data exchanged by the application of each node N is written in the data area.

  The physical destination address is a physical address indicating a direct transmission destination of non-real time data. The logical destination address is a logical address indicating the final destination of non-real time data. The logical destination address is, for example, a group address that is valid only in each of the groups G1 to G4. Note that the physical destination address and the logical destination address include a physical address and a logical address indicating the transmission source of the non-real time data. The data amount is a data amount of non-real time data.

  Returning to the description of FIG. 7, the setting unit 702 sets a route with another node N selected from another node N group. Here, the other node N is another relay node JN selected from another node N group. A route is a virtual link VL of a virtual network constructed on the network 310.

  Specifically, for example, the setting unit 702 first identifies other relay nodes N4, N7, and N10 with reference to the relay node information in the address table 500 shown in FIG. Then, the setting unit 702 sets virtual links VL1-2 to VL1-4 (see FIG. 3) that connect the node N and the other relay nodes N4, N7, and N10.

  The virtual link VL (path) is used to collectively transfer non-real time data between nodes N belonging to different groups G. The virtual link VL is set using an existing virtual network technology such as an overlay network. However, the specific setting method of the virtual link VL is omitted here because it is an existing technology. Further, the virtual link VL set between the own node N and another relay node JN is managed using, for example, the destination / transfer destination correspondence table 900 shown in FIG.

  FIG. 9 is an explanatory diagram (part 1) of a specific example of the destination / transfer destination correspondence table. In FIG. 9, the destination / transfer destination correspondence table 900 stores a destination address and a transfer destination in association with each other. Here, the destination address is a logical destination address designated as the final destination of non-real time data. Here, the group address of the node N is set as the destination address.

  A transfer destination is a transfer destination of non-real-time data. Here, the physical address of the group node GN in the own group G or the virtual link VL (identifier thereof) with the relay node JN of the other group G is set as the transfer destination. The destination / transfer destination correspondence table 900 is created, for example, according to the following procedure and stored in a storage device such as the RAM 403.

  First, the setting unit 702 refers to the address table 500 and sets the group address of the group node GN among the nodes N1 to N12 as the destination address of the destination / transfer destination correspondence table 900. Next, the setting unit 702 refers to the group identification table 600 illustrated in FIG. 6 and identifies the group G from the destination address of each group node GN.

  Here, when the specified group G is the self group G, the setting unit 702 refers to the address table 500 and sets a physical address corresponding to the destination address as the transfer destination. On the other hand, when the identified group G is another group G, the setting unit 702 refers to the address table 500 and identifies the relay node JN of the group G. Then, the setting unit 702 sets the virtual link VL with the identified relay node N as a transfer destination.

  Returning to the description of FIG. 7, the associating unit 703 associates the non-real time data received from the group node GN of the own group G with the set virtual link VL. Specifically, for example, the associating unit 703 refers to the queue correspondence table 1000 shown in FIG. 10 and puts the received non-real time data into the corresponding data queue. Here, the contents stored in the queue correspondence table 1000 will be described.

  FIG. 10 is an explanatory diagram of an example of the contents stored in the queue correspondence table. In FIG. 10, the queue correspondence table 1000 stores the transfer destination and the queue ID in association with each other. Here, the transfer destination is a transfer destination of non-real time data. The queue ID is an identifier for identifying a data queue that holds non-real time data.

  For example, when the destination address (logical destination address) of the received non-real-time data is “172.xxx.3.3”, the transfer destination is “VL1-3” (see the destination / transfer destination correspondence table 900). . Therefore, the received non-real time data enters the data queue Q4 (see the queue correspondence table 1000).

  Returning to the description of FIG. 7, the receiving unit 701 receives non-real-time data from another relay node JN via the virtual link VL. Specifically, for example, the reception unit 701 receives non-real time data from the relay node N4 of the group G2 via the virtual link VL1-1.

  The associating unit 703 associates the non-real time data received from the other relay node JN with the virtual link VL. Specifically, for example, the associating unit 703 refers to the queue correspondence table 1000 and puts the received non-real time data into the corresponding data queue.

  For example, when the destination address (logical destination address) of the non-real-time data is “172.xxx.1.2”, the transfer destination is “0000000002” (see the destination / transfer destination correspondence table 900). Therefore, the non-real time data enters the data queue Q1 (see the queue correspondence table 1000).

  The transmission unit 704 transmits non-real time data sequentially determined as a transmission target from the received data group to the other relay node JN via the virtual link VL. The transmission unit 704 transmits non-real time data sequentially determined as a transmission target from the received data group to the group node GN of the own group G.

  Here, the data group is, for example, a set of non-real time data held in the data queues Q1 to Q5. The non-real time data to be transmitted is non-real time data that is sequentially determined by a determination unit 709 described later and stored in the buffer. The buffer is, for example, a transmission buffer of NIC (Network Interface Card).

  The transmission process of the non-real time data of the transmission unit 704 is periodically executed at a constant time interval (for example, 5 [ms (milliseconds)]). Specifically, a transmission task is activated at regular time intervals, and non-real time data stored in the buffer is read in the storage order and sent to the network 310.

  That is, the transmission unit 704 receives the non-real time data from the “group node GN of its own group G” and transfers it to the “other relay node JN”. Further, the transmission unit 704 receives the non-real time data from the “other relay node JN” and transfers it to the “group node GN of the own group G”. Therefore, in the following description, the non-real time data transmission process is referred to as “non-real time data transfer process”.

  The detection unit 705 detects a cycle start point for time synchronization between the nodes N1 to N12 in the network 310. Specifically, for example, the detection unit 705 measures the elapsed time since the node N is activated, and detects the time at which a preset synchronization interval has elapsed as the cycle start time. Hereinafter, the cycle for achieving time synchronization is referred to as “cycle C1”.

  The transmission unit 704 transmits the synchronization packet to the other relay node JN as a result of detecting the period start time of the period C1. Specifically, for example, the transmission unit 704 transmits a synchronization packet provided with an identifier for time synchronization to the other relay nodes N4, N7, and N10. As a result, synchronization can be established between the relay nodes JN in the network system 300.

  In addition, the transmission unit 704 transmits the synchronization packet to the group node GN of the own group G. As a result, all the nodes N in the group G can be synchronized. In addition, the detection part 705 detects the reception time of a synchronous packet as a period start time of the period C1, when a synchronization packet is received from another relay node JN before the detection of the period start time of the period C1.

  The non-real time data transfer process is periodically executed with the period start point of the period C1 as a base point. However, the time interval at which the non-real time data transfer process is executed is equal to or shorter than the synchronization interval of the cycle C1. Hereinafter, the cycle of the non-real time data transfer process is referred to as “cycle C2”.

  The allocation unit 706 allocates a non-real-time communication flow rate to the virtual link VL based on the transmission capability of the network 310. Here, the transmission capability of the network 310 is, for example, the capacity of the physical link of the network system 300 (communication speed per unit time). The flow rate is the amount of data that can be transmitted per unit time.

  Specifically, for example, the allocation unit 706 may calculate the flow rate allocated to each virtual link VL by dividing the capacity of the physical link by the number of virtual links VL. As an example, the capacity of the physical link of the network system 300 is 12 [Mbps]. In this case, the flow rate assigned to each of the virtual links VL1-1 to VL1-3 is 4 (= 12/3) [Mbps].

  Thereby, the capacity of the physical link can be equally allocated to the virtual links VL1-1 to VL1-3. However, when a flow rate is assigned for real-time communication between the nodes N, the total assignable flow rate is a value obtained by subtracting the real-time communication assigned flow rate from the physical link capacity.

  The flow rate allocation result is stored in, for example, the allocated flow rate / transmitted data amount correspondence table 1100 shown in FIG. Further, the allocated flow rate of the virtual link VL is transmitted by the transmission unit 704 to another relay node JN. When the receiving unit 701 receives the allocated flow rate of the virtual link VL from another relay node JN, it is stored in the allocated flow rate / transmitted data amount correspondence table 1100 and the stored content is updated.

  Also, the assigning unit 706 assigns a non-real-time communication flow rate to the group node GN of the own group G based on the transmission capability of the network 310. Specifically, for example, the allocation unit 706 may calculate the flow rate allocated to each group node GN by dividing the capacity of the physical link by the number of group nodes GN of the group G1. The flow rate allocation result is stored in, for example, the allocated flow rate / transmitted data amount correspondence table 1100 shown in FIG.

  FIG. 11 is an explanatory diagram of an example of the stored contents of the allocation flow rate / transmitted data amount correspondence table. In FIG. 11, the allocation flow rate / transmitted data amount correspondence table 1100 has an allocation flow rate and a transmission data amount for each transfer destination.

  Here, the transfer destination is a transfer destination of non-real time data. The allocated flow rate is a flow rate [byte / cycle C2] of non-real-time data that can be transmitted per unit time (here, cycle C2). The transmitted data amount is the data amount [bytes] of data transmitted to the transfer destination. Here, the data amount of the non-real time data sequentially determined as a transmission target by the determination unit 709 described later is accumulated and set.

  The allocation unit 706 allocates a flow rate to each virtual link VL (or each group node GN) according to the communication status between the own node N and another relay node JN (or group node GN). You may decide. Thereby, the allocated flow rate can be controlled in accordance with the dynamically changing communication status. However, since the technique for controlling the allocated flow rate according to the communication status is an existing technology, detailed description thereof is omitted.

  Based on the transmission capability of the network 310, the first calculation unit 707 calculates a time t required for transmission of the data sequence sequentially determined as a transmission target from the start of the non-real time data transfer processing period to the present time. To do. Here, the data string is a set of non-real time data stored in the buffer. The required time t is, for example, the time required for sending the data string to the network 310.

  Specifically, for example, the first calculation unit 707 can calculate the required time t using the following formula (1). Here, t is the required time [s (seconds)] required to transmit the data string. R is a communication speed [bps (bits per second)] per unit time of the network 310. d is the data amount [b (bit)] of the data string.

    t = d / R (1)

  Here, it is assumed that the communication speed R of the network 310 is 1 [G (giga) bps] and the data amount d of the data string is 100 [K (kilo) b]. In this case, the time t required to transmit the data string is “t = 100 [Kb] / 1 [Gbps] = 100 [μs (microseconds)]”.

  The determination unit 708 determines whether or not the time when the calculated required time t has elapsed from the current time coincides with the period start time of the real-time communication that is periodically executed between any of the nodes N of the plurality of node groups. Determine whether. Here, real-time communication between the nodes N is periodically executed with a time interval (for example, 125 [μs]) shorter than the transfer process, with the period start point of the period C1 (or period C2) as a base point. Hereinafter, the cycle of real-time communication is expressed as “cycle C3”.

  Specifically, for example, the elapsed time T2 obtained by adding the required time t to the elapsed time T1 from the cycle start time of the cycle C1 (or cycle C2) to the current time is “0” or “cycle C3. It is determined whether or not it is “a constant multiple of”.

  Note that “T2 = 0” is “T1 = 0” and “t = 0”, that is, the current time is the start time of the cycle of the cycle C2, and the data to be transmitted is undecided. That is, the determination unit 708 determines whether or not the time when the required time t has elapsed from the current time coincides with the cycle start time of the real-time communication.

  Here, when the elapsed time T2 is “0” or “a constant multiple of the period C3”, the determination unit 708 determines that it coincides with the period start point of the real-time communication. On the other hand, when the elapsed time T2 is not “0” or “a constant multiple of the period T3”, the determination unit 708 determines that it does not coincide with the period start point of the real-time communication.

  FIG. 12 is an explanatory diagram illustrating an example of coincidence determination with a real-time communication cycle start time. In FIG. 12, each point on the time axis ST <b> 1 represents a period start point of real-time communication that is periodically executed between the nodes N. Points a1 and b1 on the time axes ST2 and ST3 are the cycle start time of the cycle C2, points a2 and b2 are the current time (elapsed time T1), and points a3 and b3 are the time when the required time t has elapsed from the current time (elapsed time). T2).

  In FIG. 12 (1), a time point a3 when the required time t has elapsed from the current time point a2 coincides with a real-time communication cycle start time point. In this case, the determination unit 708 determines that the time when the required time t has elapsed from the current time coincides with the cycle start time of the real-time communication.

  In FIG. 12, (2), the time point b3 when the required time t has elapsed from the current time b2 does not coincide with the real-time communication cycle start time. In this case, the determination unit 708 determines that the time when the required time t has elapsed from the present time does not coincide with the period start time of the real-time communication.

  Returning to the description of FIG. 7, the determination unit 709 determines non-real-time data to be transmitted from the received data group based on the determined determination result. Here, the data group is a set of non-real time data received from the group node GN or another relay node JN.

  Hereinafter, the determination unit 709 will be described in detail for each of (i) a case where it is determined that it matches the cycle start time of the real time communication and (ii) a case where it is determined that it does not match the cycle start time of the real time communication. An example of typical processing contents will be described.

  First, in the case of (i), the determination unit 709 specifies a path that collides with a path of real-time communication between the nodes N among the virtual links VL set between the other relay nodes JN. Here, the path that collides with the real-time communication path is the virtual link VL set up with the relay node JN of the group G to which the node N that is the communication destination of the real-time communication belongs. Note that the node N that is the communication source and communication destination of real-time communication can be specified from the real-time communication allocation flow rate table 1300 shown in FIG. 13, for example.

  FIG. 13 is an explanatory diagram (part 1) illustrating an example of the storage contents of the real-time communication allocation flow rate table. In FIG. 13, the real-time communication allocation flow rate table 1300 includes items of allocation ID, allocation flow rate, transmission source address, and destination address, and by setting information in each item, real-time communication information 1300-1 to 1300. -4 is stored as a record.

  Here, the allocation ID is an identifier for identifying the allocated flow rate allocated to the real-time communication between the nodes N. The assigned flow rate is a flow rate [bytes] assigned to real-time communication between the nodes N. The transmission source address is an address of the node N that is a communication source of real-time communication. Here, the system address and group address of the node N that is the communication source of the real-time communication are described as the transmission source address.

  The destination address is an address of the node N that is a communication destination of real-time communication. Here, the system address and group address of the node N that is the communication destination of real-time communication are described as the destination address. The specific processing content for creating the real-time communication allocation flow rate table 1300 will be described later.

  Specifically, for example, the determination unit 709 refers to the real-time communication allocation flow rate table 1300 and identifies the group address of the node N that is the communication destination of real-time communication. Here, the group addresses “172.xxx.1.2”, “172.xxx.2.2”, “172.xxx.3.2”, and “172.xxx.3” of the nodes N2, N5, N8, and N9. .3 "is specified.

  Next, the determination unit 709 refers to the destination / transfer destination correspondence table 900 and specifies a transfer destination from the specified group address. Here, the transfer destinations “0000000002 (physical address)”, “VL1-2”, and “VL1-3” are specified. The transfer destination (route) specified here is a transfer destination (route) that collides with real-time communication between the nodes N.

  That is, the transfer destinations “0000000002 (physical address)”, “VL1-2”, and “VL1-3” suppress transmission of non-real-time data in order to avoid collision with real-time communication between nodes N. Should be the forwarding destination. Therefore, for example, by using the transmittable destination list 1400 shown in FIG. 14, whether or not non-real time data can be transmitted for each transfer destination is managed.

  FIG. 14 is an explanatory diagram (part 1) of an example of the contents stored in the transmittable destination list. In FIG. 14, the transmission destination list 1400 has information (transmission availability) indicating whether or not non-real time data can be transmitted for each transfer destination.

  Here, when the transmission permission / inhibition is “permitted”, it indicates that the non-real time data can be transmitted to the transfer destination. On the other hand, if the transmission availability is “No”, it indicates that the non-real time data cannot be transmitted to the transfer destination. Note that the transmission permission / inhibition is “possible” in the initial state.

  Specifically, for example, the determination unit 709 changes the transmission permission / non-permission of the transfer destination that collides with the real-time communication between the nodes N from “permitted” to “not permitted”. Here, whether or not the specified transfer destinations “0000000002 (physical address)”, “VL1-2”, and “VL1-3” can be transmitted is changed from “permitted” to “not permitted”.

  Thereafter, the second calculation unit 710 calculates the total assigned flow rate by accumulating the assigned flow rate assigned to the real-time communication between the nodes N. Specifically, for example, the second calculation unit 710 refers to the real-time communication allocated flow rate table 1300 to calculate the total allocated flow rate obtained by accumulating the allocated flow rates of the allocations R1 to R4. Here, the total allocated flow rate is 5800 [bytes].

  Then, the extracting unit 711 extracts non-real-time data whose transfer destination transmission permission is “permitted” from the received data group. Specifically, for example, the extraction unit 711 refers to the transmittable destination list 1400 and identifies a transfer destination whose transmission is possible. Here, the transfer destinations “0000000003 (physical address)” and “VL1-4” are specified.

  Next, the extraction unit 711 refers to the queue correspondence table 1000 and identifies the identified data queue of the transfer destination. Here, the data queue Q2 of the transfer destination “0000000003 (physical address)” and the data queue Q5 of the transfer destination “VL1-4” are specified. Then, the extraction unit 711 extracts non-real time data from the specified data queue (here, the data queues Q2 and Q5).

  The determination unit 709 determines the extracted non-real time data as non-real time data to be transmitted. Here, if non-real time data having a data amount larger than the total allocated flow rate of real time communication is set as a transmission target, there is a possibility of collision with real time communication in the next cycle C3. Therefore, the determination unit 709 determines non-real time data whose data amount is equal to or less than the calculated total allocated flow rate among the extracted non-real time data as non-real time data to be transmitted.

  However, there is a possibility of data loss if non-real time data exceeding the allocated flow rate that can be transmitted per unit time (see FIG. 11) is transmitted. For this reason, the determination unit 709 does not determine non-real time data for a transfer destination exceeding the allocated flow rate as a transmission target.

  When non-real time data to be transmitted does not exist, the determination unit 709 determines invalid data to be discarded in the switch SW1 directly connected to the own node N as a transmission target. As invalid data, for example, an Ethernet frame with a pause 0 [s] can be used.

  At this time, the determination unit 709 may determine invalid data corresponding to the total allocated flow rate of real-time communication as a transmission target. Note that the non-real time data to be transmitted does not exist means that there is no non-real time data to be extracted or the non-real time data that has been extracted is a non-real time data that is less than or equal to the total allocated flow rate. This is the case when there is no real-time data.

  In addition, when there are a plurality of non-real time data equal to or less than the total allocated flow rate, the determining unit 709 transmits non-real time data having the maximum data amount from among the plurality of non-real time data. You may decide to. Thereby, non-real time data can be transmitted efficiently.

  Next, in the case of (ii), the second calculation unit 710 calculates the remaining time from the time when the required time t has elapsed from the current time until the start of the real-time communication cycle. Specifically, for example, first, the second calculation unit 710 calculates an elapsed time T3 from the cycle start point of the cycle C1 (or cycle C2) to the cycle start point of real-time communication immediately after the current time point.

  Then, the second calculation unit 710 can calculate the remaining time until the real time communication cycle start time is reached by subtracting the elapsed time T2 from the calculated elapsed time T3. The elapsed time T2 is an elapsed time obtained by adding the required time t to the elapsed time T1 from the cycle start time of the cycle C1 (or cycle C2) to the present time.

  The second calculation unit 710 calculates the amount of data that can be transmitted in the remaining time based on the transmission capability of the network 310. Specifically, for example, the second calculation unit 710 can calculate the amount of data that can be transmitted in the remaining time by using the following equation (2). Here, X is the amount of data [b] that can be transmitted in the remaining time, (T3-T2) is the remaining time [s], and R is the communication speed [bps] per unit time of the network 310.

    X = (T3-T2) × R (2)

  Here, the remaining time (T3-T2) is set to 20 [μs], and the communication speed R of the network 310 is set to 1 [Gbps]. In this case, the data amount X that can be transmitted in the remaining time (T3−T2) is “X = 20 [μs] × 1 [Gbps] = 20 [M (mega) b]”.

  Then, the extraction unit 711 extracts non-real time data that is less than or equal to the amount of data that can be transmitted in the remaining time from the non-real time data group. Specifically, for example, the extraction unit 711 extracts any non-real time data that is less than or equal to the amount of data that can be transmitted in the remaining time from the data queues Q1 to Q5.

  The determination unit 709 determines the extracted non-real time data as data to be transmitted. At this time, when there are a plurality of extracted non-real time data, the determination unit 709 determines the non-real time data having the maximum data amount as non-real time data to be transmitted from the plurality of data. Also good.

  When there is no non-real time data to be transmitted, the determination unit 709 determines invalid data to be discarded in the switch SW1 directly connected to the own node N as a transmission target. At this time, the determination unit 709 may determine invalid data corresponding to the amount of data that can be transmitted in the remaining time as a transmission target.

  Further, the non-real time data determined as the transmission target is stored in the buffer. Then, the non-real time data stored in the buffer is sent to the network 310 in the storage order when the transmission task of the transmission unit 704 is activated.

  In this way, by suppressing transmission of non-real time data that collides with real-time communication, collision between real-time communication and non-real-time communication can be avoided. When there are a plurality of real-time communications executed in the network system 300, the non-real-time data to be transmitted is determined in consideration of the start time of each real-time communication.

  The determined determination result is reflected in the allocation flow rate / transmitted data amount correspondence table 1100 shown in FIG. Specifically, in the allocation flow rate / transmitted data amount correspondence table 1100, the transmitted data amount of the transfer destination of the non-real time data determined as the transmission target is updated. More specifically, the data amount of the non-real time data determined as the transmission target is added to the current transmitted data amount.

  The receiving unit 701 receives a real-time communication flow rate allocation request from the group node GN or another relay node JN. The flow rate allocation request includes a request flow rate for real-time communication, a communication source address, and a communication destination address. Here, the required flow rate is a flow rate required for real-time communication between the nodes N.

  The communication source address is a transmission source address of data related to real time communication (hereinafter referred to as “real time data”), and is, for example, a system address of the node N of the transmission source. The communication destination address is a destination address of real-time data, for example, a system address of the destination node N. In the following description, the source of the flow rate allocation request (group node GN or other relay node JN) is referred to as “forward source”.

  The determination unit 712 determines whether the requested flow rate can be allocated as a result of receiving the flow rate allocation request for real-time communication. Specifically, for example, first, the determination unit 712 refers to the real-time communication allocation flow rate table 1300, and sums the allocation flow rate (hereinafter referred to as “node allocation”) whose destination address matches the destination address included in the flow rate allocation request. The total flow rate s ”is calculated.

Thereafter, the determination unit 712 calculates the provisional node allocation flow total s ′ by adding the requested flow included in the flow allocation request to the node allocation flow total s. Then, the determination unit 712 determines whether or not the temporary node allocation flow total s ′ is equal to or less than the maximum data amount F _MAX that can be transmitted during the real-time communication cycle C3.

Here, when the temporary node allocation flow total s ′ is larger than the maximum data amount F _MAX , the transmission unit 704 transmits a flow allocation response indicating that allocation is impossible to the forward source. That is, the amount of data transmitted to the node N that is the communication destination of real-time communication is prevented from exceeding the maximum data amount F _MAX that can be transmitted during the period C3, and the data on the network 310 is avoided. Prevent loss.

On the other hand, when the temporary node allocation flow rate total s ′ is equal to or less than the maximum data amount F _MAX , the determination unit 712 refers to the address table 500 and specifies a group address corresponding to the destination system address included in the flow rate allocation request. . Then, the determination unit 712 refers to the group identification table 600 and identifies the group G corresponding to the identified group address.

  Thereafter, the determination unit 712 determines whether or not the identified group G is the own group G. Here, in the case of the own group G, the determining unit 712 refers to the real-time communication allocation flow rate table 1300 and refers to the total of the real-time communication allocation flow rates of the other group G and the destination is the own group G (hereinafter, “Assigned flow rate total S”) is calculated.

Thereafter, the determination unit 712 adds the required flow rate to the total allocated flow rate S to calculate the temporary allocated flow total S ′. Then, the determination unit 712 determines whether or not the total temporarily allocated flow rate S ′ is equal to or less than the maximum data amount F _MAX that can be transmitted during the real-time communication cycle C3.

Here, when the temporary allocation flow rate total S ′ is larger than the maximum data amount F _MAX , the transmission unit 704 transmits a flow rate allocation response indicating that allocation is impossible to the forward source. That is, the loss of data on the network 310 is prevented by avoiding that the amount of data transmitted to the own group G exceeds the maximum data amount F _MAX that can be transmitted during the period C3.

On the other hand, when the temporary allocation flow rate total S ′ is equal to or less than the maximum data amount F _MAX , the determination unit 712 determines to allocate the required flow rate for real-time communication. Then, the determination unit 712 sets the requested flow rate, the transmission source address, and the destination address included in the flow rate allocation request in the real time communication allocation flow rate table 1300, and registers new real time communication information as a record. Note that the group addresses of the source and destination node N are specified from the address table 500 using the source and destination system addresses. Finally, the transmission unit 704 transmits a flow rate allocation response indicating that allocation is possible to the forward source.

  On the other hand, when the specified group G is another group G, the determination unit 712 refers to the real-time communication allocation flow rate table 1300, and allocates the real-time communication allocation flow rate of the source group G and the destination of the other group G. The total S is calculated. The specified group G is the group G specified from the group address corresponding to the destination system address included in the flow rate allocation request.

Thereafter, the determination unit 712 adds the required flow rate to the total allocated flow rate S to calculate the temporary allocated flow total S ′. Then, the determination unit 712 determines whether or not the total temporarily allocated flow rate S ′ is equal to or less than the maximum data amount F _MAX that can be transmitted during the real-time communication cycle C3.

Here, when the temporary allocation flow rate total S ′ is larger than the maximum data amount F _MAX , the transmission unit 704 transmits a flow rate allocation response indicating that allocation is impossible to the forward source. That is, the amount of data transmitted from the own group G to the other group G is prevented from exceeding the maximum data amount F _MAX that can be transmitted during the period C3, and data loss on the network 310 is prevented. .

On the other hand, when the temporary allocation flow total S ′ is equal to or less than the maximum data amount F _MAX , the transmission unit 704 transmits a flow allocation request to the identified relay node JN of the group G with reference to the address table 500. Thereafter, when the receiving unit 701 receives a flow rate allocation response indicating that allocation is not possible from the relay node JN, the transmission unit 704 transmits a flow rate allocation response indicating that allocation is not possible to the forward source.

  On the other hand, when the reception unit 701 receives a flow rate allocation response indicating that allocation is possible from the relay node JN, the determination unit 712 determines that the requested flow rate is allocated to real-time communication. Then, the determination unit 712 sets the requested flow rate, the transmission source address, and the destination address included in the flow rate allocation request in the real time communication allocation flow rate table 1300, and registers new real time communication information as a record.

  This avoids flow allocation exceeding the transmission capacity of the network 310 in response to a flow allocation request for real-time communication, prevents data from being discarded during communication of the switches SW1 to SW4, etc. Communication quality can be ensured.

(Functional configuration of group node GN)
Next, a functional configuration of the group node GN will be described. FIG. 15 is a block diagram showing a functional configuration of the group node. In FIG. 15, the group node GN includes a receiving unit 1501, a detecting unit 1502, a transmitting unit 1503, an assigning unit 1504, a first calculating unit 1505, a determining unit 1506, a determining unit 1507, and an extracting unit 1508. And a second calculation unit 1509.

  Specifically, each function unit (reception unit 1501 to second calculation unit 1509) executes the program stored in the storage device such as the ROM 402, the RAM 403, the magnetic disk, and the optical disk illustrated in FIG. Or the function is realized by the I / F 404. In the following description, when a specific example of each functional unit (receiving unit 1501 to second calculating unit 1509) of the group node GN is described, the node N2 in the network system 300 is assumed as the group node GN.

  The receiving unit 1501 receives various data. The various data are, for example, a synchronization packet, a flow allocation result of non-real time communication, a flow allocation result of real time communication, a flow allocation response of real time communication, real time data, and non-real time data.

  The detection unit 1502 detects a cycle start point for time synchronization between the nodes N1 to N12 in the network 310. Specifically, as a result of receiving the synchronization packet from relay node JN, detection unit 1502 detects the reception time of the synchronization packet as the period start time of period C1.

  The transmission unit 1503 transmits real-time data. Specifically, for example, the transmission unit 1503 transmits real-time data sequentially determined as a transmission target from the data group to other group nodes GN. Here, the data group is, for example, a set of real-time data and non-real-time data held in the data queue. Further, real-time data to be transmitted is sequentially determined by a determination unit 1507 described later and stored in a buffer.

  The transmission unit 1503 transmits non-real time data. Specifically, for example, the transmission unit 1503 transmits non-real time data sequentially determined as a transmission target from the data group to the relay node JN. Here, the non-real-time data to be transmitted is sequentially determined by a determination unit 1507 described later and stored in the buffer.

  The transmission processing of the real-time data and the non-real-time data by the transmission unit 1503 is periodically executed in the cycle C2 with the cycle start point of the cycle C1 as a base point. Specifically, a transmission task is activated at a time interval of the period C2, and real-time data or non-real-time data stored in the buffer is read in the storage order and sent to the network 310. At this time, since the data string in the buffer is shaped like, for example, the data string 210 shown in FIG. 2, real-time data is transmitted at time intervals of the period C3.

  Note that the destination of the non-real time data is set with reference to the destination / transfer destination correspondence table 1600 shown in FIG. 16, for example. FIG. 16 is an explanatory diagram (part 2) of a specific example of the destination / transfer destination correspondence table. In FIG. 16, the destination / transfer destination correspondence table 1600 stores destination addresses and transfer destinations in association with each other.

  Here, the destination address is a logical destination address designated as the final destination of non-real time data. Here, the group address of the node N is set as the destination address. A transfer destination is a transfer destination of non-real-time data. Here, the physical address of the group node GN or the relay node JN is set as the transfer destination.

  For example, when non-real time data is transmitted to the group node N3 of the own group G1, the physical address of the group node N3 is set as the physical destination address. On the other hand, when transmitting non-real-time data to the group node N5 of the other group G2, the physical address of the relay node N1 is set as the physical destination address, and the group address of the group node N5 is set as the logical destination address.

  Returning to the description of FIG. 15, the allocation unit 1504 allocates the received flow allocation result of non-real-time communication to the path connecting the group node GN and the relay node JN. Specifically, for example, the allocation unit 1504 holds the physical link that connects the node and the relay node JN in association with the flow rate allocation result.

  In addition, the transmission unit 1503 transmits a flow rate allocation request for real-time communication to the relay node JN. Here, the flow rate allocation request for real-time communication is a flow rate allocation request required for real-time communication using the node N as a communication source. The real-time communication flow rate allocation request includes information indicating the requested flow rate, the system address of the real-time communication source (local node), and the system address of the real-time communication destination (other group node GN). ing.

  When a flow rate allocation response including information indicating that the requested flow rate can be allocated is received as a result of the transmission of the real time communication flow rate allocation request, for example, it is stored in the real time communication allocated flow rate table 1700 shown in FIG. The When the flow allocation result of real time communication is received, it is stored in the real time communication allocated flow table 1700.

  FIG. 17 is an explanatory diagram (part 2) of an example of the stored contents of the real-time communication allocation flow rate table. In FIG. 17, a real-time communication allocation flow rate table 1700 has items of allocation ID, allocation flow rate, transmission source address, and destination address, and records information on real-time communication information 1600-1 by setting information in each item. Remember as.

  The real-time communication allocation flow rate table 1700 manages real-time communication information in which the own node N is a communication source or communication destination of real-time communication. Here, real-time communication information 1700-1 with the node N2, which is its own node N, as the communication destination (destination) is stored.

  In addition, when allocation is impossible as a result of receiving the flow rate allocation response, the group node GN may autonomously determine whether or not the real-time communication required flow rate may be reduced. For example, if the requested flow rate for real-time communication can be reduced, there is no problem if the requested flow rate for real-time communication is greater than the actual required flow rate or the bit rate can be reduced when transferring video data, etc. There are cases. In this case, the transmission unit 1503 retransmits to the relay node JN a flow rate allocation request in which the request flow rate for real-time communication is less than the previous request flow rate.

  Based on the transmission capability of the network 310, the first calculation unit 1505 is required for transmission of a data string sequentially determined as a transmission target from the start of the cycle of transmission processing of real-time data and non-real-time data. Time t is calculated. Specifically, for example, the first calculation unit 1505 can calculate the required time t required to transmit the data string using the above equation (1).

  The determination unit 1506 determines whether or not the time when the calculated required time t has elapsed from the current time coincides with the period start time of the real-time communication that is periodically executed between any of the nodes N of the plurality of node groups. Determine whether. Specifically, for example, the elapsed time T2 obtained by adding the required time t to the elapsed time T1 from the cycle start point of the cycle C2 (or cycle C1) to the current time is “0” or “cycle C3”. It is determined whether or not it is “a constant multiple of”.

  The determination unit 1507 determines data to be transmitted from the data group based on the determined determination result. Here, the data group is a set of real-time data and non-real-time data. Specifically, for example, the determination unit 1507 refers to the real-time communication allocation flow rate table 1700 and specifies the group address of the node N that is the communication source of real-time communication. Here, the group address “172.xxx.2.3” of the node N6 is specified.

  Next, the determination unit 1507 compares the identified group address with the group address of the own node N, and determines whether or not the communication source of the real-time communication is the own node N. Here, when the communication source is the own node N, the determination unit 1507 determines the real time data at the head of the data queue of the communication destination (destination) of the real time communication as a transmission target.

  Here, the communication source of the real-time communication is not the own node N. In this case, the determination unit 1507 refers to the real-time communication allocation flow rate table 1700 and specifies the group address of the node N that is the communication destination of real-time communication. Here, the group address “172.xxx.1.2” of the node N2 is specified.

  Next, the extraction unit 1508 refers to the destination / transfer destination table correspondence 1500 and specifies the transfer destination from the specified group address. The transfer destination (route) specified here is a transfer destination (route) that collides with real-time communication between the nodes N. That is, the transfer destination specified here is a transfer destination for which transmission of non-real time data should be suppressed in order to avoid a collision with the real time communication between the nodes N. Therefore, for example, by using a transmittable destination list 1800 shown in FIG. 18, whether or not non-real time data can be transmitted for each transfer destination is managed.

  FIG. 18 is an explanatory diagram (part 2) of an example of the contents stored in the transmittable destination list. In FIG. 18, the transmission destination list 1800 has information (transmission availability) indicating whether or not non-real time data can be transmitted for each transfer destination.

  Specifically, for example, the determination unit 1507 changes the transmission availability of the transfer destination that collides with the real-time communication between the nodes N from “available” to “not available”. However, since the node N2 is its own node N, the transfer destination is not specified here. Therefore, there is no change in whether transmission is possible in the transmission destination list 1800.

  Thereafter, the second calculation unit 1509 determines the relay node JN based on the data amount (transmitted data amount) of the data string determined as the transmission target from the cycle start time of the cycle C1 (or cycle C2) to the present time. The amount of data that can be transmitted is calculated using a path connecting between and. Specifically, for example, the second calculation unit 1509 subtracts the transmitted data amount from the allocated flow rate allocated to the route connecting to the relay node JN and transmits the data amount (hereinafter, “residual allocation”). Flow rate)).

  Then, the extraction unit 1508 extracts, from the data group, data that indicates whether or not the transmission destination can be transmitted and that is equal to or less than the calculated remaining allocated flow rate. Specifically, for example, the extraction unit 1508 refers to the destination / transfer destination correspondence table 1600 and identifies a transfer destination whose transmission is possible. Here, transfer destinations “0000000001 (physical address)” and “0000000003 (physical address)” are specified.

  Next, the extraction unit 1508 extracts non-real-time data in the specified transfer destination data queue. The determination unit 1507 determines the extracted non-real time data as non-real time data to be transmitted. At this time, the determination unit 1507 determines, among the extracted non-real time data, non-real time data whose data amount is equal to or less than the calculated remaining allocated flow rate as non-real time data to be transmitted.

  In addition, when there are a plurality of non-real time data below the remaining allocated flow rate, the determination unit 1507 determines the non-real time data with the largest data amount as non-real time data to be transmitted from the plurality of data. May be.

  If there is no non-real time data to be transmitted, the determination unit 1507 determines invalid data to be discarded in the switch SW1 directly connected to the own node N as a transmission target. At this time, the determination unit 1507 may determine invalid data corresponding to the remaining allocated flow rate as a transmission target.

  Next, in the case of (ii), the second calculation unit 1509 calculates the remaining time from when the required time t has elapsed from the current time until the start time of real-time communication. Specifically, for example, first, the second calculation unit 1509 calculates the elapsed time T3 from the cycle start time of the cycle C2 (or cycle C1) to the start time of real-time communication immediately after the current time.

  Then, the second calculation unit 1509 can calculate the remaining time until the start time of the real-time communication is reached by subtracting the elapsed time T2 from the calculated elapsed time T3. The elapsed time T2 is an elapsed time obtained by adding the required time t to the elapsed time T1 from the cycle start point of the cycle C2 (or cycle C1) to the present time.

  The second calculation unit 1509 calculates the amount of data that can be transmitted in the remaining time based on the transmission capability of the network 310. Specifically, for example, the second calculation unit 1509 can calculate the amount of data that can be transmitted in the remaining time using the above equation (2).

  Then, the extraction unit 1508 extracts, from the data group, data equal to or less than the data amount that can be transmitted in the remaining time. Specifically, for example, the extraction unit 1508 extracts any non-real time data that is less than or equal to the amount of data that can be transmitted in the remaining time from the data queue.

  The determination unit 1507 determines the extracted non-real time data as non-real time data to be transmitted. At this time, when there are a plurality of extracted non-real time data, the determination unit 1507 determines the non-real time data having the maximum data amount as non-real time data to be transmitted from the plurality of data. Also good.

  However, there is a possibility of data loss if non-real time data exceeding the allocated flow rate that can be transmitted per unit time is transmitted. For this reason, the determination unit 1507 does not determine non-real time data exceeding the allocated flow rate as a transmission target even if the amount of data is less than or equal to the amount of data that can be transmitted in the remaining time. The determined determination result is stored in a storage area such as the RAM 403 as the transmitted data amount.

  If there is no non-real time data to be transmitted, the determination unit 1507 determines invalid data to be discarded in the switch SW1 directly connected to the own node N as a transmission target. At this time, the determination unit 1507 may determine invalid data for the amount of data that can be transmitted in the remaining time as a transmission target.

  Note that the real-time data and non-real-time data determined as transmission targets are stored in the buffer. The real time data and non-real time data stored in the buffer are sent to the network 310 in the order of storage when the transmission task is activated. In this way, by suppressing transmission of non-real time data that collides with real-time communication, collision between real-time communication and non-real-time communication can be avoided.

(Communication procedure of relay node JN)
Next, a communication processing procedure of the relay node JN will be described. However, it is assumed that the virtual link VL between the relay nodes JN is set in advance. FIG. 19 is a flowchart illustrating an example of a communication processing procedure of the relay node. In the flowchart of FIG. 19, first, the detection unit 705 determines whether one cycle of the cycle C1 has elapsed since the start of the node N (step S1901).

  If one cycle has not elapsed (step S1901: NO), the receiving unit 701 determines whether a synchronization packet has been received from another relay node JN (step S1902). If no synchronization packet has been received (step S1902: NO), the process returns to step S1901.

  On the other hand, when the synchronization packet is received (step S1902: Yes), the detection unit 705 detects the reception time of the synchronization packet as the cycle start time of the cycle C1 (step S1903). If one cycle has elapsed in step S1901 (step S1901: Yes), the current time point is detected as the cycle start point of cycle C1 (step S1904).

  Then, the transmission unit 704 transmits a synchronization packet to another relay node JN (step S1905). Next, the detection unit 705 determines whether or not the current time is the cycle start time of the cycle C1 (step S1906). Specific processing contents for determining whether or not the current time point is the cycle start time point of the cycle C1 will be described later.

  Here, when it is not the period start time of the period C1 (step S1906: No), the process proceeds to step S1914. On the other hand, in the case of the cycle start time of the cycle C1 (step S1906: Yes), the allocating unit 706 allocates the non-real-time communication flow rate to the group node GN in the own group G based on the transmission capability of the network 310 (step S1906: Yes). S1907). Then, the allocation unit 706 registers the flow rate allocation result of the group node GN in the allocated flow rate / transmitted data amount correspondence table 1100 (step S1908).

  Also, the allocation unit 706 allocates a non-real-time communication flow rate to the virtual link VL with another relay node JN based on the transmission capability of the network 310 (step S1909). Then, the allocation unit 706 registers the flow rate allocation result of the virtual link VL in the allocation flow rate / transmitted data amount correspondence table 1100 (step S1910).

  Thereafter, the transmission unit 704 transmits the synchronization packet and the flow rate allocation result of the group node GN to the group node GN (step S1911). In addition, the transmission unit 704 transmits the synchronization packet and the flow rate allocation result of the other relay node JN to the other relay node JN (step S1912).

  Next, the detection unit 705 executes a cycle start time correction process for correcting the cycle start time of the cycle C1 (step S1913). Then, the transmission unit 704 executes a transfer control process for transferring the non-real time data to the other relay node JN via the virtual link VL (step S1914).

  Thereafter, the detection unit 705 determines whether or not there is an instruction to stop the network system 300 (step S1915). Note that an instruction to stop the network system 300 is received from an external computer device via the network 310, for example.

  Here, when there is no stop instruction (step S1915: No), the process waits until the current time becomes the cycle start time of the cycle C2 (step S1916: No), and when the cycle start time of the cycle C2 comes (step S1916: Yes). ), The process returns to step S1906. On the other hand, when there is a stop instruction (step S1915: Yes), a series of processes according to this flowchart is ended.

  Here, an example of specific processing contents for determining whether or not the current time point is the cycle start time point of the cycle C1 in step S1906 will be described. Here, the ratio between the time interval of the cycle C1 and the time interval of the cycle C2 is “cycle C1: cycle C2 = N: 1”. In addition, the number of loops in the cycle C2 is n (where n is an initial value “n = 0”). In this case, when the remainder of the value obtained by dividing (n + 1) by N is 0, the detection unit 705 determines that the current time is the cycle start time of the cycle C1.

<Cyclic start time correction process>
Next, a specific processing procedure of the cycle start time correction process in step S1913 shown in FIG. 19 will be described. FIG. 20 is a flowchart illustrating an example of a specific processing procedure of the cycle start time correction processing in step S1913.

  In the flowchart of FIG. 20, first, the receiving unit 701 determines whether or not a synchronization packet and a flow rate allocation result have been received from another relay node JN (step S2001). If the synchronization packet and the flow rate allocation result have not been received (step S2001: NO), the process proceeds to step S2006.

  On the other hand, when the synchronization packet and the flow rate allocation result are received (step S2001: Yes), the allocation unit 706 registers the flow rate allocation result of the virtual link VL in the allocation flow rate / transmitted data amount correspondence table 1100 (step S2002). Then, the detection unit 705 determines whether or not there is a deviation at the cycle start time of the cycle C1 from the reception time of the synchronization packet (step S2003).

  If there is no deviation at the cycle start time (step S2003: No), the process proceeds to step S2005. On the other hand, when there is a deviation at the cycle start time (step S2003: Yes), the detection unit 705 detects the reception time of the synchronization packet as the cycle start time of the cycle C1 (step S2004).

  Thereafter, the receiving unit 701 determines whether or not the synchronization packet and the flow rate allocation result have been received from all the other relay nodes JN (step S2005). If the synchronization packet and the flow rate allocation result are received (step S2005: YES), the process proceeds to step S1914 shown in FIG. On the other hand, when the synchronization packet and the flow rate allocation result have not been received (step S2005: No), the allocation unit 706 determines whether or not a predetermined time has elapsed from the initial determination process in step S2001 (step S2001). S2006).

  Here, when the fixed time has not elapsed (step S2006: No), the process returns to step S2001. On the other hand, when the predetermined time has elapsed (step S2006: Yes), the allocation unit 706 allocates the synchronization packet and the flow rate allocation result to “0” as the allocation flow rate of the virtual link VL with another relay node JN that has not received it. The data is registered in the flow rate / transmitted data amount correspondence table 1100 (step S2007), and the process proceeds to step S1914 shown in FIG.

  In step S2007, if the allocated flow rate of the virtual link VL is registered as “0” continuously for the specified number of times, it is determined that another relay node JN connected by the virtual link VL has left the network system 300. It may be.

<Transfer control procedure>
Next, a specific processing procedure of the transfer control process in step S1914 shown in FIG. 19 will be described. 21 and 22 are flowcharts showing an example of a specific processing procedure of the transfer control processing in step S1914.

  In the flowchart of FIG. 21, first, the first calculation unit 707 initializes the transmitted data amount of each transfer destination in the allocation flow rate / transmitted data amount correspondence table 1100 with “0” (step S2101). Also, the first calculation unit 707 initializes the transmittable destination list 1400 (step S2102).

  Thereafter, the first calculation unit 707 accumulates the transmitted data amount of each transfer destination in the allocation flow rate / transmitted data amount correspondence table 1100, and sequentially determines the transmission target from the cycle start time to the present time of the cycle C2. A data amount d of the obtained data string is calculated (step S2103). Then, the first calculation unit 707 calculates the required time t required to transmit the data string using the above equation (1) (step S2104).

  Next, the determination unit 708 determines whether or not the time when the required time t has elapsed from the current time coincides with the cycle start time of the real-time communication cycle C3 (step S2105). Here, when it coincides with the cycle start point of the cycle C3 (step S2105: Yes), the determination unit 709 refers to the real-time communication allocation flow rate table 1300 and the group address of the node N that is the communication destination of the real-time communication Is identified (step S2106).

  Thereafter, the determination unit 709 refers to the destination / transfer destination table correspondence 800 and specifies the transfer destination from the specified group address (step S2107). Then, the determination unit 709 changes the transmission permission / non-permission of the specified transfer destination in the transmission destination list 1400 from “permitted” to “not permitted” (step S2108).

  Next, the second calculation unit 710 refers to the real-time communication allocation flow rate table 1300 to calculate the total allocation flow of real-time communication by accumulating the allocation flow allocated to the real-time communication between the nodes N. (Step S2109). Then, the determining unit 709 substitutes the calculated total allocated flow rate into MaxSendN (step S2110), and the process proceeds to step S2114 in FIG. Note that MaxSendN is a variable.

  In step S2105, if it does not coincide with the cycle start time of cycle C3 (step S2105: No), the second calculation unit 710 sets the cycle start time of cycle C3 from the time when the required time t has elapsed from the present time. The remaining time until is calculated (step S2111).

  Thereafter, the second calculation unit 710 calculates the amount of data that can be transmitted in the remaining time using the above equation (2) (step S2112). Then, the determining unit 709 substitutes the calculated data amount into MaxSendN (step S2113), and the process proceeds to step S2114 in FIG. Note that MaxSendN is a variable.

  In the flowchart of FIG. 22, first, the extraction unit 711 refers to the transmission destination list 1400 to identify a transfer destination whose transmission permission is “permitted” (step S2114). Next, the extraction unit 711 refers to the queue correspondence table 1000 to identify the identified data queue of the transfer destination (step S2115).

  Thereafter, the extraction unit 711 extracts non-real time data of MaxSendN or less from the specified transfer destination data queue (step S2116). Then, the determining unit 709 determines the extracted non-real time data as non-real time data to be transmitted (step S2117). However, the determination unit 709 does not determine non-real time data for a transfer destination exceeding the allocated flow rate as a transmission target.

  In step S2117, if there is no non-real time data to be transmitted, the determination unit 709 determines invalid data to be discarded in the switch SW1 directly connected to the own node N as a transmission target. At this time, the determination unit 709 determines invalid data for MaxSendN as a transmission target.

  Next, the determining unit 709 accumulates the data amount of the non-real-time data determined as the transmission target in the transmitted data amount of the transfer destination in the allocation flow rate / transmitted data amount correspondence table 1100 (step S2118). Then, the determining unit 709 calculates the total transmitted data amount by accumulating the transmitted data amount of each transfer destination in the allocation flow rate / transmitted data amount correspondence table 1100 (step S2119).

  Thereafter, the determining unit 709 determines whether or not the calculated total transmitted data amount is less than the transmittable amount in the period C2 (step S2120). If it is less than the transmittable amount (step S2120: Yes), the process returns to step S2102 shown in FIG. On the other hand, if the amount exceeds the transmittable amount (step S2120: No), the process proceeds to step S1915 shown in FIG.

  Thereby, in network system 300, collision between real-time communication and non-real-time communication can be avoided. Further, since only the transmission of non-real time data colliding with real time communication is suppressed, it is possible to suppress a decrease in throughput of the entire network 310.

(Communication processing procedure of group node GN)
Next, a communication processing procedure of the group node GN will be described. FIG. 23 is a flowchart illustrating an example of a communication processing procedure of the group node. In the flowchart of FIG. 23, first, the receiving unit 1501 determines whether or not a synchronization packet has been received from the relay node JN (step S2301).

  Here, it waits for the reception of the synchronization packet (step S2301: No), and when it is received (step S2301: Yes), the detection unit 1502 detects the reception time of the synchronization packet as the period start time of the period C1 ( Step S2302). Thereafter, the detection unit 1502 determines whether or not the current time point is the cycle start point of the cycle C1 (step S2303).

  Here, when it is not the period start time of the period C1 (step S2303: No), the process proceeds to step S2308. On the other hand, in the case of the cycle start time of cycle C1 (step S2303: Yes), the receiving unit 1501 determines whether a synchronization packet and a flow rate allocation result are received from the relay node JN (step S2304).

  Here, waiting for receiving the synchronization packet and the flow rate allocation result (step S2304: No), if received (step S2304: Yes), the allocation unit 1504 displays the received non-real-time communication flow rate allocation result. The node N and the relay node JN are assigned to the path connecting them (step S2305).

  Next, the detection unit 1502 determines whether or not there is a deviation at the cycle start time of the cycle C1 from the reception time of the synchronization packet (step S2306). If there is no deviation at the cycle start time (step S2306: NO), the process proceeds to step S2308. On the other hand, when there is a deviation at the cycle start time (step S2306: Yes), the detection unit 1502 detects the reception time of the synchronization packet as the cycle start time of the cycle C1 (step S2307).

  Then, the transmission unit 1503 executes transmission control processing for transmitting real-time data and / or non-real-time data (step S2308). Thereafter, the detection unit 1502 determines whether or not there is an instruction to stop the network system 300 (step S2309).

  Here, when there is no stop instruction (step S2309: No), it waits until the current time becomes the cycle start time of the cycle C2 (step S2310: No), and when the cycle start time of the cycle C2 is reached (step S2310: Yes). ), The process returns to step S2303. On the other hand, when there is a stop instruction (step S2309: Yes), a series of processing according to this flowchart is ended.

<Transmission control processing procedure>
Next, a specific processing procedure of the transmission control process in step S2308 shown in FIG. 23 will be described. 24 and 25 are flowcharts showing an example of a specific processing procedure of the transmission control processing in step S2308.

  In the flowchart of FIG. 24, first, the first calculation unit 1505 initializes the transmitted data amount of the path connecting the own node N and the relay node JN with “0” (step S2401). In addition, the first calculation unit 1505 initializes the transmittable destination list 1400 (step S2402).

  Thereafter, the first calculation unit 1505 identifies the data amount d of the data string sequentially determined as the transmission target from the start of the cycle of the cycle C2 to the present time from the transmitted data amount (step S2403). Then, the first calculation unit 1505 calculates the required time t required to transmit the data string using the above equation (1) (step S2404).

  Next, the determination unit 1506 determines whether or not the time when the required time t has elapsed from the current time coincides with the cycle start time of the real-time communication cycle C3 (step S2405). Here, when it does not coincide with the cycle start time of the cycle C3 (step S2405: No), the second calculation unit 1509 performs the remaining from the time when the required time t has elapsed from the current time until the cycle start time of the cycle C3. Is calculated (step S2406).

  Thereafter, the second calculation unit 1509 calculates the amount of data that can be transmitted in the remaining time using the above equation (2) (step S2407), and the process proceeds to step S2416. On the other hand, in step S2405, when it coincides with the cycle start time of cycle C3 (step S2405: Yes), the determination unit 1507 refers to the real-time communication allocation flow rate table 1300 to refer to the node N that is the communication source of real-time communication. Is specified (step S2408).

  Then, the determining unit 1507 compares the identified group address with the group address of the own node N to determine whether or not the communication source of the real-time communication is the own node N (step S2409). Here, in the case of the own node N (step S2409: Yes), the determination unit 1507 determines the real time data at the head of the data queue of the communication destination of the real time communication as a transmission target (step S2410).

  Then, the determining unit 1507 accumulates the data amount of the real-time data determined as the transmission target in the transmitted data amount (step S2411), and proceeds to step S2421 shown in FIG.

  On the other hand, when it is not the own node N (step S2409: No), the determining unit 1507 identifies the group address of the node N that is the communication destination of the real-time communication (step S2412). Thereafter, the determination unit 1507 refers to the destination / transfer destination correspondence table 1600 to specify the transfer destination from the specified group address (step S2413).

  Then, the determination unit 1507 changes the transmission permission / non-permission of the specified transfer destination in the transmission destination list 1800 from “available” to “not” (step S2414). Next, the second calculating unit 1509 calculates the remaining allocated flow rate by subtracting the transmitted data amount from the allocated flow rate allocated to the path connecting the node N and the relay node JN (step S2415).

  Then, the determination unit 1507 substitutes the smaller of the remaining allocated flow rate or the transmittable data amount for MaxSendN (step S2416), and proceeds to step S2417 in FIG.

  In the flowchart of FIG. 25, first, the extraction unit 1508 refers to the transmission destination list 1800 to identify a transfer destination whose transmission permission is “permitted” (step S2417). Next, the extraction unit 1508 extracts non-real time data less than or equal to MaxSendN from the identified transfer destination data queue (step S2418).

  Then, the determining unit 1507 determines the extracted non-real time data as non-real time data to be transmitted (step S2419). Next, the determining unit 1507 accumulates the data amount of the non-real time data determined as the transmission target in the transmitted data amount (step S2420).

  If there is no non-real time data to be transmitted in step S2419, the determination unit 1507 determines invalid data to be discarded in the switch SW directly connected to the own node N as a transmission target. At this time, the determination unit 1507 determines invalid data for MaxSendN as a transmission target.

  Thereafter, the determining unit 1507 determines whether the transmitted data amount is less than the allocated flow rate (step S2421). If the flow rate is less than the allocated flow rate (step S2421: YES), the process returns to step S2402 shown in FIG. On the other hand, when the flow rate is equal to or higher than the allocated flow rate (step S2421: NO), the process proceeds to step S2309 shown in FIG.

  Thereby, in network system 300, collision between real-time communication and non-real-time communication can be avoided. Further, since only the transmission of non-real time data colliding with real time communication is suppressed, it is possible to suppress a decrease in throughput of the entire network 310.

<Flow allocation request processing procedure for real-time communication>
Next, a flow rate allocation request processing procedure for real-time communication of the group node GN will be described. FIG. 26 is a flowchart illustrating an example of a flow rate allocation request processing procedure for real-time communication. In the flowchart of FIG. 26, first, the transmission unit 1503 transmits a flow rate allocation request for real-time communication to the relay node JN (step S2601).

  Thereafter, the reception unit 1501 waits for reception of a flow rate allocation response from the relay node JN (step S2602: No), and if received (step S2602: Yes), whether or not allocation is possible by the group node GN. Judgment is made (step S2603).

  Here, when allocation is possible (step S2603: Yes), the group node GN registers the real-time communication information as a new record in the real-time communication allocation flow rate table 1700 (step S2604), and a series of processes according to this flowchart. Exit.

  On the other hand, when allocation is impossible (step S2603: No), the group node GN determines whether or not the requested flow rate for real-time communication may be reduced (step S2605). Here, when the required flow rate cannot be reduced (step S2605: No), the series of processes according to this flowchart is terminated.

  On the other hand, when the required flow rate may be reduced (step S2605: Yes), the group node GN calculates a new real-time communication required flow rate (step S2606), and returns to step S2601.

<Flow allocation response processing procedure for real-time communication>
Next, the flow rate allocation response processing procedure of the real-time communication of the relay node JN will be described. 27 and 28 are flowcharts illustrating an example of a flow rate allocation response processing procedure for real-time communication. In the flowchart of FIG. 27, first, the receiving unit 701 determines whether or not a real-time communication flow rate allocation request has been received from the forward source (step S2701).

  Here, it waits for reception of a flow allocation request for real-time communication (step S2701: No), and when received (step S2701: Yes), the determination unit 712 refers to the real-time communication allocation flow table 1300. Then, the node allocation flow total s is calculated (step S2702).

Next, the determination unit 712 adds the requested flow rate included in the flow rate allocation request to the calculated node allocation flow rate total s to calculate the temporary node allocation flow rate total s ′ (step S2703). Then, the determination unit 712 determines whether or not the total temporary node allocation flow rate s ′ is equal to or less than the maximum data amount F _MAX that can be transmitted during the real-time communication cycle C3 (step S2704).

Here, if the temporary node allocation flow rate total s ′ is larger than the maximum data amount F _MAX (step S2704: No), the transmission unit 704 transmits a flow rate allocation response indicating that allocation is impossible to the forward source (step S2705). A series of processing by this flowchart is complete | finished.

On the other hand, when the temporary node allocation flow total s ′ is equal to or less than the maximum data amount F _MAX (step S2704: Yes), the determination unit 712 refers to the address table 500 and corresponds to the destination system address included in the flow allocation request. The group address to be specified is specified (step S2706).

  Then, the determining unit 712 refers to the group identification table 600 and identifies the group G corresponding to the identified group address (step S2707). Thereafter, the determination unit 712 determines whether the identified group G is the own group G (step S2708).

  Here, in the case of the own group G (step S2708: Yes), the determination unit 712 refers to the real-time communication allocation flow rate table 1300, and allocates real-time communication in which the transmission source is the other group G and the destination is the own group G. The total flow S is calculated (step S2709).

Thereafter, the determination unit 712 adds the requested flow rate to the calculated allocated flow rate total S to calculate the temporary allocated flow rate total S ′ (step S2710). Then, the determination unit 712 determines whether or not the temporary allocation flow total S ′ is equal to or less than the maximum data amount F _MAX that can be transmitted during the real-time communication cycle C3 (step S2711).

Here, when the temporary allocation flow total S ′ is larger than the maximum data amount F _MAX (step S2711: No), the process proceeds to step S2705. On the other hand, when the temporary allocation flow total S ′ is equal to or less than the maximum data amount F _MAX (step S2711: Yes), the determination unit 712 registers new real-time communication information in the real-time communication allocation flow rate table 1300 (step S2712). .

  Then, the transmission unit 704 transmits a flow rate allocation response indicating that allocation is possible to the forward source (step S2713). Further, the transmission unit 704 transmits real-time communication information to other relay nodes JN excluding the forward source and the group node GN of the own group G (step S2714). Thereby, the newly registered real-time communication information is shared.

  In step S2708, when the identified group is another group G (step S2708: No), the process proceeds to step S2715 shown in FIG.

  In the flowchart shown in FIG. 28, first, the determination unit 712 refers to the real-time communication allocation flow rate table 1300 to calculate the total real-time communication allocation flow rate S of which the transmission source is the own group G and the destination is the other group G. (Step S2715).

Thereafter, the determination unit 712 adds the requested flow rate to the calculated allocated flow rate total S to calculate the temporary allocated flow rate total S ′ (step S2716). Then, the determination unit 712 determines whether or not the temporary allocation flow total S ′ is equal to or less than the maximum data amount F _MAX that can be transmitted during the real-time communication cycle C3 (step S2717).

If the temporary allocation flow rate total S ′ is larger than the maximum data amount F _MAX (step S2717: No), the transmission unit 704 transmits a flow rate allocation response indicating that allocation is impossible to the forward source (step S2718). A series of processes according to the flowchart ends.

On the other hand, when the temporary allocation flow total S ′ is equal to or less than the maximum data amount F _MAX (step S2717: Yes), the transmission unit 704 transmits a flow allocation request to the identified relay node JN of the group G (step S2719). Thereafter, the reception unit 701 waits to receive a flow rate allocation response from another relay node JN (step S2720: No).

  When the flow rate allocation response is received (step S2720: Yes), the determination unit 712 determines whether the flow rate allocation response indicates that allocation is possible (step S2721). Here, when allocation is impossible (step S2721: No), it transfers to step S2718.

  On the other hand, when allocation is possible (step S2721: Yes), the determination unit 712 registers new real-time communication information in the real-time communication allocation flow rate table 1300 (step S2722). Finally, the transmission unit 704 transmits a flow rate allocation response indicating that allocation is possible to the forward source (step S2723), and the series of processes according to this flowchart is terminated.

  This avoids flow allocation exceeding the transmission capacity of the network 310 in response to a flow allocation request for real-time communication, prevents data from being discarded during communication of the switches SW1 to SW4, etc. Communication quality can be ensured.

  As described above, according to the present embodiment, it is possible to transfer non-real-time data between groups G using the virtual link VL that connects between the relay nodes JN selected from each group G. Thereby, the non-real-time data between the groups G can be integrated into one and the collision with the real-time communication can be reduced.

  Further, according to the present embodiment, the transmission time of the data to be stored next is predicted from the time t required for sending the data string stored in the buffer to the network 310, and the real-time communication is started. It can be determined whether or not it coincides with the time.

  As a result, when it coincides with the start time of real-time communication, non-real-time data of a path that does not collide with real-time communication can be determined as a transmission target. On the other hand, if it does not coincide with the start time of real-time communication, non-real-time data that is less than or equal to the amount of data that can be transmitted before the start time of real-time communication can be determined as a transmission target.

  Thereby, it is possible to suppress only the transmission of non-real time data colliding with the real time communication, and to suppress a decrease in the throughput of the entire network 310. In addition, since the data string is shaped and stored in the buffer so as not to collide with the real-time communication, transmission timing control becomes unnecessary.

  For example, when a moving image shot by a camera is transferred by real-time communication with a very short cycle C3 (for example, 125 [μs]) by a timer process (interrupt process), the resources of the CPU 401 are used up only for the transmission process. It is assumed that other processes cannot be executed. However, according to the present embodiment, since it is not necessary to control the transmission timing in accordance with the extremely short cycle C3, even if each node N is not equipped with a high-performance CPU 401, a collision with real-time communication can be achieved. It can be avoided.

  Further, according to the present embodiment, even when there is no appropriate non-real time data to be transmitted, the invalid data discarded in the switch SW is used as the transmission target, thereby colliding with real time communication. You can format data strings that don't.

  Further, according to the present embodiment, the type of protocol for performing real-time communication is not limited. Therefore, it is not necessary to set a protocol for real-time communication that is generally required (for example, RTP: Real-time Transport Protocol), and it is possible to reduce the load on the operator who constructs the system.

  Further, according to the present embodiment, since the network system 300 can be constructed using a plurality of switches SW, the flexibility and expandability of the system configuration can be improved. Furthermore, according to the present embodiment, communication quality can be ensured even when real-time communication between nodes N is performed over a plurality of switches SW. Further, in a multi-stage switch-connected local network in which two types of communication are mixed, an isochronous transfer with an extremely short period can be executed without delay by avoiding a collision between the two types of communication.

  Therefore, according to the disclosed node, communication program, and communication method, it is possible to ensure the communication quality of real-time communication even in an environment where high-load non-real-time communication is performed. At this time, since it is possible to control so that real-time communication and non-real-time communication do not collide by a single physical communication medium connecting the nodes N, it is not necessary to prepare a communication medium for each communication, It is possible to prevent the complexity and cost of the system configuration from increasing.

  In the communication method described above, there is a topology restriction that the switch SW that is not directly connected to the relay node JN cannot be connected adjacently. For example, when a switch SW that connects switches SW such as a core switch and a router switch is installed, a switch SW that is not directly connected to the relay node JN is connected adjacently. In this case, data may collide between the switches SW that are not directly connected to the relay node JN.

  Therefore, the relay node JN may be installed in the switch SW that connects the switches SW. Specifically, the relay node JN is installed in the switch SW that connects the switch SW directly connected to the relay node JN and the switch SW not directly connected to the relay node JN.

  FIG. 29 is a system configuration diagram (part 2) of the network system. In the network system 2900, small groups SG1 to SG15 formed by nodes (not shown) directly connected to the individual switches SW1 to SW16 are communicably connected via the network 310.

  The large group BG1 is formed by the small groups SG1 to SG5, the large group BG2 is formed by the small groups SG6 to SG10, and the large group BG3 is formed by the small groups SG11 to SG15. Here, the switches SW5, SW10, and SW15 are core switches that connect the switches. Therefore, relay nodes 2910, 2920, and 2930 are installed in the switches SW5, SW10, and SW15.

  In the network system 2900, non-real time communication exceeding the large groups BG1 to BG3 is performed by transmitting non-real time data to a large group relay node (for example, the relay nodes 2910, 2920, 2930) once as communication between small groups. To do. Moreover, the communication between large groups can avoid the collision with real-time communication and non-real-time communication by implementing the communication processing procedure similar to the relay nodes between small groups. In this way, it is possible to cope with a large-scale system by preventing adjacent switches having no relay node from being adjacent to each other.

  The communication method described in this embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. The communication program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. The communication program may be distributed via a network such as the Internet.

  The following additional notes are disclosed with respect to the embodiment described above.

(Supplementary note 1) The first node group including the first node and the second node and the second node group including the third node and the fourth node are connected to each other via the network. The first node performing communication and second communication, comprising:
Setting means for setting a communication path via the network between the first node and the third node;
Determining means for determining data to be transmitted from the data received from the second node, which is data related to the second communication performed between the second node and the fourth node;
Transmitting means for transmitting the determined data to be transmitted to the third node via the communication path;
Based on the transmission capability of the network, calculation means for calculating a time required for transmission of the transmission target data from the start time of transmission processing periodically executed by the transmission means to the present time;
Determining whether or not the time when the required time has elapsed from the current time coincides with the start time of the first communication periodically executed between the second node and the fourth node Means,
With
The determining means determines the data to be transmitted from the received data based on the determination result determined by the determining means.
A node characterized by that.

(Supplementary Note 2) When it is determined that the time coincides with the start time of the first communication, the second node group including the fourth node serving as a communication destination of the first communication is determined from the received data. An extraction means for extracting arbitrary data not destined for the node of 3;
The node according to claim 1, wherein the determining unit determines the data extracted by the extracting unit as the data to be transmitted.

(Additional remark 3) The said determination means is a 1st which is performed between the said 2nd node and the said 4th node when the said required time passes from the said present time among the data extracted by the said extraction means. The node according to appendix 2, wherein data that is equal to or less than a data amount of data related to communication is determined as the data to be transmitted.

(Supplementary Note 4) When it is determined that the calculation means does not coincide with the start time of the first communication, the first communication is started from the time when the required time has elapsed from the current time based on the transmission capability of the network. The amount of data that can be transmitted in the remaining time until the start time of
The node according to appendix 2 or 3, wherein the extraction unit extracts data equal to or less than the data amount calculated by the calculation unit from the received data.

(Additional remark 5) The said determination means determines the data discarded in the relay apparatus of the said network to the said transmission target data, when the said transmission target data does not exist among the received data, It is characterized by the above-mentioned. The node according to any one of appendices 1 to 4.

(Appendix 6) The transmission means includes:
Any one of appendices 1 to 5, wherein the transmission target data sequentially determined from data received from the third node via the communication path is transmitted to the second node. Node described in.

(Appendix 7) Nodes in a network in which a plurality of node groups are connected via a relay device,
Transmission means for transmitting data sequentially determined as a transmission target from among the data group to one node selected from a node group including the own node among the plurality of node groups;
Calculation means for calculating a time required for transmission of the data sequence sequentially determined as the transmission target from the start time of transmission processing periodically executed by the transmission means to the current time based on the transmission capability of the network; ,
It is determined whether or not the time when the required time calculated by the calculation means has elapsed from the current time coincides with the start time of communication processing periodically executed between any one of the plurality of node groups. A determination means;
A determination unit that determines the data to be transmitted from the data group based on the determination result determined by the determination unit;
A node characterized by comprising:

(Supplementary note 8) The first node group including the first node and the second node and the second node group including the third node and the fourth node are connected to each other via the network. A communication program to be executed by the computer of the first node that executes communication and second communication,
A receiving step of receiving data related to second communication performed between the second node and the fourth node from the second node;
A setting step of setting a communication path through the network between the first node and the third node;
Based on the transmission capability of the network, the transmission target sequentially determined from the start point of the transmission process periodically executed to transmit the data sequentially determined as the transmission target from the received data to the present time A calculation step for calculating the time required for data transmission;
Determining whether or not the time when the required time has elapsed from the current time coincides with the start time of the first communication periodically executed between the second node and the fourth node Process,
A determination step of determining the transmission target data from the received data based on the determination result determined by the determination step;
A communication program for causing a computer to execute.

(Supplementary note 9) The first node group including the first node and the second node and the second node group including the third node and the fourth node are connected to each other via the network. A computer of the first node that performs communication and second communication;
A receiving step of receiving data related to second communication performed between the second node and the fourth node from the second node;
A setting step of setting a communication path through the network between the first node and the third node;
Based on the transmission capability of the network, the transmission target sequentially determined from the start point of the transmission process periodically executed to transmit the data sequentially determined as the transmission target from the received data to the present time A calculation step for calculating the time required for data transmission;
Determining whether or not the time when the required time has elapsed from the current time coincides with the start time of the first communication periodically executed between the second node and the fourth node Process,
A determination step of determining the transmission target data from the received data based on the determination result determined by the determination step;
The communication method characterized by performing.

300, 2900 Network system 310 Network 701, 1501 Reception unit 702 Setting unit 703 Association unit 704, 1503 Transmission unit 705, 1502 Detection unit 706, 1504 Allocation unit 707, 1505 First calculation unit 708, 1506 Determination unit 709, 1507 Determination Unit 710, 1509 Second calculation unit 711, 1508 Extraction unit 712 Judgment unit Group node GN
Relay node JN
N1-N12 node SW1-SW5 switch

Claims (7)

Between the first node group including the first node and the second node and the second node group including the third node and the fourth node; The first node executing the communication of
Setting means for setting a communication path via the network between the first node and the third node;
Determining means for determining data to be transmitted from the data received from the second node, which is data related to the second communication performed between the second node and the fourth node;
Transmitting means for transmitting the determined data to be transmitted to the third node via the communication path;
Based on the transmission capability of the network, calculation means for calculating a time required for transmission of the transmission target data from the start time of transmission processing periodically executed by the transmission means to the present time;
Determining whether or not the time when the required time has elapsed from the current time coincides with the start time of the first communication periodically executed between the second node and the fourth node Means,
With
The determining means determines the data to be transmitted from the received data based on the determination result determined by the determining means.
A node characterized by that. When it is determined that it coincides with the start time of the first communication, from the received data, the third node of the second node group including the fourth node serving as a communication destination of the first communication is determined. It has extraction means to extract any data that is not a destination,
The node according to claim 1, wherein the determining unit determines the data extracted by the extracting unit as the data to be transmitted.   The determining means includes data of the first communication executed between the second node and the fourth node when the required time has elapsed from the current time out of the data extracted by the extracting means. The node according to claim 2, wherein data that is equal to or less than a data amount is determined as the data to be transmitted. When it is determined that the calculation means does not coincide with the start time of the first communication, the start time of the first communication is determined from the time when the required time has elapsed from the current time based on the transmission capability of the network. Calculate the amount of data that can be sent in the remaining time until
4. The node according to claim 2, wherein the extraction unit extracts data equal to or less than the data amount calculated by the calculation unit from the received data. 5.   The determination unit determines data to be discarded in the relay device of the network as the data to be transmitted when the data to be transmitted does not exist among the received data. The node as described in any one of -4. Between the first node group including the first node and the second node and the second node group including the third node and the fourth node; A communication program to be executed by the computer of the first node that executes communication of
A receiving step of receiving data related to second communication performed between the second node and the fourth node from the second node;
A setting step of setting a communication path through the network between the first node and the third node;
Based on the transmission capability of the network, the transmission target sequentially determined from the start point of the transmission process periodically executed to transmit the data sequentially determined as the transmission target from the received data to the present time A calculation step for calculating the time required for data transmission;
Determining whether or not the time when the required time has elapsed from the current time coincides with the start time of the first communication periodically executed between the second node and the fourth node Process,
A determination step of determining the transmission target data from the received data based on the determination result determined by the determination step;
A communication program for causing a computer to execute. Between the first node group including the first node and the second node and the second node group including the third node and the fourth node; A computer of the first node that performs communication of
A receiving step of receiving data related to second communication performed between the second node and the fourth node from the second node;
A setting step of setting a communication path through the network between the first node and the third node;
Based on the transmission capability of the network, the transmission target sequentially determined from the start point of the transmission process periodically executed to transmit the data sequentially determined as the transmission target from the received data to the present time A calculation step for calculating the time required for data transmission;
Determining whether or not the time when the required time has elapsed from the current time coincides with the start time of the first communication periodically executed between the second node and the fourth node Process,
A determination step of determining the transmission target data from the received data based on the determination result determined by the determination step;
The communication method characterized by performing.

Images